Q2 2026 uptime report: 99.98%

Over the three months, we saw two incidents that briefly affected customers.\n\nIncident #1 — April 12, 06:12-06:53 EAT. A misbehaving image-processing queue on one shared-hosting node caused elevated CPU and slower response times. We isolated the node, drained traffic, and rate-limited the job type. Total customer-visible impact: 41 minutes of degraded performance on approximately 8% of shared-hosting customers.\n\nIncident #2 — June 3, 22:04-22:21 EAT. A cache warm-up bug caused webmail authentication to take up to 8 seconds. Immediate mitigation was a cache reset; the fix landed within a week.\n\nWhat we improved: added synthetic monitoring for all mail-serving endpoints, tightened per-queue rate limits on shared hosting, and doubled our on-call rotation coverage during EAT peak hours (18:00-00:00).\n\nThe Q3 report will follow the same format.